DE102015002559A1 - Stabilizing optical frequency combs - Google Patents

Abstract

The invention relates to methods for operating a laser device (1) with which an optical frequency comb can be stabilized, the frequencies of whose modes can be described by the formula fm = m × frep + f0, where frep is a mode spacing, f0 is an offset frequency and m is a natural Number is. At least one signal (S1, S2, S3, S4) is determined, which correlates with an actual value of a degree of freedom (F), wherein the degree of freedom (F) is a linear combination of the offset frequency f0 and the mode spacing frep of the frequency comb. Based on the signal, the actual value of the degree of freedom (F) is set with a first control unit (10) in a predetermined capture area (F) of a second control unit (40). As soon as the capture range (ΔF range) of the second control unit (40) has been reached, the second control unit (40) is activated and the actual value is regulated to a desired value (FSoll) with the aid of the second control unit (40).

Description

The invention relates to the field of optical frequency combs. As explained below, the inventive method for stabilizing an optical frequency comb can be used.

As is known, an optical frequency comb can be generated by providing a short or ultrashort laser pulse. Such a laser pulse may have a pulse duration in the range of picoseconds or femtoseconds. However, larger or smaller pulse durations are conceivable, for example in the range of attose to microseconds.

In the frequency domain, at least one laser pulse generated in a resonator can be represented as a frequency comb. 2 shows such a representation as a frequency comb, wherein the optical intensity I is plotted against the frequency f. The frequency comb comprises modes at discrete optical frequencies f _m . An envelope of the intensity profile can lie within the amplification bandwidth of a laser medium generating the laser pulse. The width of this envelope is inversely proportional to the pulse duration of the laser pulse.

As is known, the frequencies of the modes of an optical frequency comb can be described generally by the formula f _m = m × f _rep + f ₀ , where m is a natural number and f _rep and f _{0 are} the unit of a frequency. Like from this formula and 2 As can be seen, the frequencies of adjacent modes of the optical frequency comb the distance f _rep , which is referred to as the mode spacing of the frequency comb. If the frequency comb corresponds to a laser pulse circulating in a resonator, the mode spacing f _rep corresponds to the pulse repetition frequency (= repetition rate) of the oscillator, that is to say to the inverse of the circulating time of the pulse in the oscillator.

The modes of the frequency comb are normally not exactly at an integer multiple of f _rep . As from the above formula and 2 can be seen, the frequency comb can be shifted by an offset frequency f ₀ . However, it is also conceivable that f _{0 is} equal to zero and thus the modes of the frequency comb lie at integer multiples of f _rep . For a laser pulse circulating in a resonator, the presence of the offset frequency f ₀ may be due to the fact that the group velocity of a pulse circulating in the oscillator differs from the phase velocity of the individual modes of the pulse.

It obvious to the skilled person that the description of the modes of the frequency comb by the formula f _m = m × f _rep + f ₀ is an idealized representation. The modes of a real frequency comb may have a finite width in frequency space.

Optical frequency combs are used for example in the field of spectroscopy, in particular the spectroscopy of electronic transitions in atoms, or for very precise frequency measurements. For these applications, it is important to be able to stabilize a frequency comb. "Stabilization of a frequency comb" in the sense of the invention means a stabilization of the position of at least one of the modes of the frequency comb. Active stabilization may be necessary because the mode spacing f _rep and the offset frequency f _{0 can be} extremely sensitive to external influences. For a laser pulse in a resonator even a slight change in the resonator length and thus the repetition rate leads to a change in the mode spacing f _{rep of} the frequency comb. For example, a change in the offset frequency f ₀ can lead to a change in the dispersive properties in the resonator.

In the DE 199 11 103 A1 and the DE 100 44 404 A1 is the stabilization of an optical frequency comb by controlling the two parameters offset frequency f ₀ and mode spacing f _m disclosed. The DE 199 11 103 A1 discloses changing the resonator length via a displaceable deflection mirror in order to regulate the mode spacing of the frequency comb. In addition, it is disclosed to set the offset frequency by tilting a resonator end mirror or by introducing a prism pair into the beam path of the resonator.

After providing a frequency comb, the parameters of the frequency comb (offset frequency f ₀ and mode spacing f _rep ) can be indefinite. In general, at least one of these parameters is indeterminate immediately after the frequency comb is provided. Even during the operation of the frequency comb, at least one of the parameters can be indefinite or indefinite, for example by changing a physical boundary condition.

Known control circuits for stabilizing the frequency comb can only be activated unreliably. This is because an ordinary control loop can only stabilize a parameter of the frequency comb if it is already (randomly) in an area defined by the characteristics of the control loop. Otherwise, the control loop typically begins to oscillate, running toward one end of its control range, or driving the frequency comb based on a noise signal.

It is therefore an object of the present invention to provide a method for operating a laser device with which a frequency comb can be easily and reliably stabilized.

This object is achieved by the method according to claim 1. The dependent claims indicate advantageous embodiments of the invention.

A frequency comb can be generated, for example, via a femtosecond laser (Fs laser), in particular an Fs fiber laser. The frequency comb can be generated, for example, in a resonator, in particular a microresonator, or by means of an optical parametric oscillator (OPO). The generation of the frequency comb by modulation of continuous wave lasers (EOM combs, combination of phase and frequency modulation) or by optical rectification is conceivable. The use of a difference frequency comb is also conceivable. The frequency comb can also be provided in a different way. It is also conceivable that the frequency comb used was subsequently changed by AOMs or EOMs.

According to the invention, an optical frequency comb is provided which has a multiplicity of modes whose frequencies can be described by the formula f _m = m × f _rep + f ₀ . This frequency comb should now be stabilized. For this purpose, at least one signal is determined which correlates with an actual value of a degree of freedom F of the frequency comb. The degree of freedom F of the frequency comb is according to the invention a linear combination of the offset frequency f ₀ and the mode spacing f _{rep of} the frequency comb. The degree of freedom F can thus be expressed by the following formula: F = v × f _rep + w × f ₀ + c, where v and w are real numbers and c is the unit of a frequency. Exclude here can only be the case that v and w are simultaneously 0. Explicitly included within the scope of the present invention are the cases F = f _rep (ie v = 1, w = c = 0) and F = f ₀ (w = 1, v = c = 0). The degree of freedom F in each case characterizes the frequency comb, or properties of the frequency comb. For a complete description of a frequency comb at least one degree of freedom F may be necessary.

According to the invention, a first input signal, which is contained in the quantity of the at least one signal, is transmitted to a first control unit. Advantageously, a plurality of first input signals, which are each contained in the quantity of the at least one signal, can be transmitted to the first control unit.

Based on the transmitted at least one first input signal, according to the invention, the actual value of the degree of freedom is set in a predetermined capture range of a second control unit with the first control unit. This is a step preceding the actual stabilization of the actual value, which ensures that the corresponding control loop is effective. The actual stabilization is then carried out by the second control unit.

If the actual value is in the capture range of the second control unit, the second control unit is activated. Whether the actual value lies in the capture range can be determined, for example, by re-measuring the actual value of the degree of freedom.

If the second control unit is activated, the actual value is controlled with the aid of this to a predetermined desired value. This stabilizes the frequency comb. Depending on the specific degree of freedom F selected, a setpoint suitable for a specific application can be defined. The actual value of the degree of freedom does not necessarily have to be brought into exact agreement with the desired value for successful stabilization. It may be sufficient if the actual value of the degree of freedom is kept within a specified range around the setpoint. In this case, the desired value can be selected such that the mode spacing f _{rep of} the frequency comb z. B. is stabilized in a range around 250 MHz and / or the offset frequency f ₀ is stabilized in a range between 0 and 125 MHz. The stabilization can take place in such a way that the actual value is identical with the desired value up to a few Hz, for example up to at most 2 Hz, 5 Hz, 10 Hz, 20 Hz, 50 Hz or 100 Hz or so that only deviations of the actual Values of less than 1%, less than 3%, less than 5% or less than 10% are allowed. If the degree of freedom F corresponds to the mode spacing f _{rep of} the frequency comb, the stabilization can be carried out in such a way that the actual value is identical with the desired value except for 1 mHz or 2 mHz. If the degree of freedom F corresponds to the offset frequency f _{0 of} the frequency comb, the stabilization can be carried out such that the actual value is identical to the desired value except for 1 Hz or 2 Hz. The agreement with the desired value can be assessed within a certain integration time, for example over one or more seconds.

According to the invention, the laser device is thus operated with a two-stage process. By means of the first control unit, the actual value of the degree of freedom is brought into the predetermined capture range of the second control unit. This ensures that the second control unit is able to reliably control the actual value to the setpoint. In principle, a known system for stabilizing a frequency comb could be used as the second control unit. The method according to the invention then ensures that that the control by the second control unit can be reliably activated and is ready to operate quickly.

At least one second input signal, which is contained in the quantity of the at least one signal, that is correlated with the actual value of the degree of freedom, can be transmitted to the second control unit. Advantageously, the second control unit comprises a single second control loop or a plurality of second control loops, which controls one or more actuators based on the at least one second input signal.

It is conceivable that exactly one actuator is assigned to each second control loop. Alternatively, an actuator can also be controlled by a plurality of second control circuits. Advantageously, all second control circuits control the one or more actuators based on a single second input signal included in the set of the at least one signal. The single second input signal is thus used jointly by the second control circuits. Thus, the stabilization of the frequency comb can be simplified because the number of signals to be measured is maximally reduced. Even if only a single second input signal is used, several control stages can be implemented within the second control unit in order to implement the most efficient and fast control possible and / or to cover a larger control range. It is also conceivable that every second control loop has its own second input signal.

The second control loops can be cascaded. In this case, upstream second control circuits can bring downstream second control circuits into their control ranges or hold them there. In particular, in the cascading upstream second control loops can have a coarser setting accuracy than the downstream control loops. This can be realized by a suitable choice of the actuators or the second input signals used. Advantageously, the cascaded second control circuits access different actuators which may have different setting accuracies.

The second control circuits with the associated actuators can be designed so that they are able to reliably control the actual value of the degree of freedom to the predetermined desired value, if the actual value is in the capture range of the second control unit. The capture area may be limited by several properties of the second control unit. For example, if a piezoelectric actuator is used as the actuator of the second control unit for moving a resonator, the capture range can be limited by the limited travel of the piezoelectric actuator. It is also conceivable that an error signal used for stabilization (for example the difference between the relevant second input signal and the predetermined desired value of the degree of freedom) is greatly attenuated outside a certain range, so that no stabilization is possible. It is also conceivable that the second input signal used for stabilization is outside a detection range. This can be particularly problematic if the offset frequency f _{0 of} a frequency comb is to be controlled.

In order to bring the actual value of the degree of freedom into the capture range of the second control unit, the first control unit may comprise a plurality of first control loops. Advantageously, these are used sequentially one after the other. In a simple embodiment, it is also conceivable that the first control unit comprises only a first control loop.

The first control circuits may use the at least one first input signal to control the actual value. It is conceivable that the at least one first input signal comprises the second input signal. In this way, an already provided measuring device can be used to determine the second input signal from the first control unit. Advantageously, the at least one first input signal comprises at least one signal which is not used by the second control unit. Such a first input signal is usefully measurable over a larger range than the signals used by the second control unit. Increasing the measuring range is generally accompanied by a certain deterioration of the measuring resolution. However, once the actual value of the degree of freedom has to be brought into the capture range of the second control unit with the first control unit, this can be accepted for the first control unit. For a signal used by the first control unit, a broad detection range is more important than a high resolution, since in this way also actual values of the degree of freedom far away from the desired value can be detected and brought into the capture range of the second control unit.

It is advantageous if the first control unit optionally activates one or more second control circuits. Thus, the control can be activated by the second control unit directly from the first control unit as soon as the actual value is in the capture range. Advantageously, several second control loops can be activated separately. Thus, in each case a situation-based best suited second control loop can be activated. It is also conceivable, all or more second control loops activate together, for example, if they are cascaded.

The first control unit may include a state machine that may be configured to activate the second control loops. The state machine may also coordinate the sequential use of the first control loops. For this purpose, the state machine can be supplied with the measured at least one signal. Using the state machine, the method according to the invention can be further automated. Thus, a reliable fully automatic activation of a frequency comb stabilization can be implemented without having to move "manually" actuators "blindly" until stabilization takes effect.

It may be advantageous if the first control unit uses and / or adjusts one or more of the actuators controlled by the second control circuits in parallel or as an alternative to the second control unit. So the number of necessary actuators can be reduced.

It is also conceivable that at least one actuator is controlled by the first control unit, which is independent of the second control unit. This actuator can have a larger adjustment range than the actuators controlled by the second control unit. As a result, actual values far away from the desired value can also be brought into the capture range of the second control unit. A reduction of the setting accuracy possibly accompanied by the larger adjustment range can generally be accepted in the first control unit.

The first and the second control unit can be designed to process at least one actuator signal, which represents the control value of an actuator.

The at least one first input signal used by the first control unit expediently has the widest possible range of validity, preferably substantially centered around the desired value. In this case, "validity range" designates the value range of the degree of freedom in which the determination of the at least one first input signal supplies a correct measured value. Advantageously, the validity range of the first input signal is greater than the capture range of the second control unit, in particular at least one and a half times, twice, three times, five times, ten times or twenty times as large. If the mode spacing f _rep for determining the first input signal is measured, the capture range may include 500 Hz (measurement by phase detector) and the validity range 10 MHz (measurement by counter). If the offset frequency f _{0 is} measured to determine the first input signal, the capture range can comprise 2 MHz and the validity range 20 MHz.

If the actual value of the degree of freedom nevertheless lies outside the validity range of the first input signal, it is expedient for the first control unit to be able to detect this with the aid of a further first input signal and then to change a manipulated value of at least one actuator until the validity range is reached again. Advantageously, the control value of the at least one actuator is changed uniformly, for example, in constant steps. This ensures that the scope is reached again. Alternatively, the manipulated variable of the actuator can be stochastically changed in order to get back into the scope as quickly as possible. Advantageously, outside the validity range, the at least one actuator is actuated as a function of one or more preceding states of the first control unit in order to bring the actual value back into the validity range as efficiently as possible. The manipulation of the manipulated variable of at least one actuator can be controlled by the state machine of the first control unit. This is advantageous because the state machine is designed anyway for driving the actuators. In addition, the process can be further automated. For this purpose, the state machine can drive a function generator which, depending on the input of the state machine, outputs a control signal (s) to one or more actuators.

To check whether the actual value is within the validity range, a power level of a beat signal can be determined. For this purpose, a laser pulse forming the optical frequency comb can be made to interfere with a reference signal of known frequency. This produces a beat whose beat frequency corresponds to the difference frequency between the frequency f _{m of} a mode of the frequency comb and the frequency of the reference signal. The optical beat signal may be converted to an electrical signal, for example via a photodiode, and then passed through a frequency filter. By evaluating the signal level, it can be determined whether the beat frequency is in a predetermined range (passband of the frequency filter). With a suitable choice of the reference signal and the frequency filter can be determined so that the actual value is within the validity range. It is also conceivable to measure with an f: 2f interferometer.

Since capacitances for determining a signal, which correlate with the actual value of the degree of freedom, can already be provided within the scope of the present invention anyway, it can be determined without great additional effort whether a successful stabilization of the degree of freedom has occurred. For this purpose, it can be determined whether the actual value lies within a certain range around the setpoint, preferably over a predetermined period of time. If a condition of successful stabilization is detected, it can be displayed or output in a message. This makes it possible to integrate the frequency comb into a higher-level automated system. The frequency comb can report to a higher-level system unit when it is ready for use by that system. Thus, it is not necessary to supervise the turn-on procedure of the frequency comb by a user.

In the following, the invention and its advantages will be further clarified by means of drawings. Showing:

1 1 is a schematic representation of a section of an exemplary laser device that can be operated with the method according to the invention;

2 a frequency-space representation of modes of an exemplary frequency comb, wherein the frequency is plotted on the horizontal axis and the light intensity on the vertical axis,

3 a schematic representation of some relevant value ranges for the inventive degree of freedom of a frequency comb, and

4 a schematic representation of at least part of a signal processing system for performing the method according to the invention according to an embodiment.

An exemplary laser device 1 , which can be operated by the method according to the invention, is in 1 shown. However, it should be noted that other laser devices can be operated with the method according to the invention, as long as the provision of an optical frequency comb is possible.

At the laser device 1 out 1 is a laser-active medium 2 on an optical axis 3 a resonator 4 intended. In the laser-active medium 2 it may, for example, be a Ti: Sa crystal. There are also other laser-active media, in particular laser crystals, such. Yb: YAG, Cr: LiSAF or Cr: Forsterite conceivable. The laser-active medium 2 comes with from one outside the resonator 4 arranged pump laser 5 generated pumped laser radiation P pumped.

The resonator 4 can have several resonator mirrors 6a . 6b . 6c . 6d include. In the illustrated embodiment, the resonator 4 a linear resonator. In this case, two resonator mirrors make up the mirrors 6a and 6b , Resonatorendspiegel. On the optical axis 3 of the resonator 4 can be between the resonator end mirrors 6a . 6b in any number of other resonator mirrors 6c . 6d be arranged. Alternatively, it is conceivable, the resonator 4 form as a ring resonator, so that the resonator 4 has no Resonatorendspiegel.

One of the resonator mirrors 6d is advantageously designed as Einkoppelspiegel, which is suitable for coupling of pumped laser radiation P. In addition, preferably a Auskoppelspiegel (in 1 the resonator mirror 6b ) for coupling out laser light from the resonator 4 intended. For improved beam guidance, it may be expedient if some resonator mirrors 6a . 6b . 6c . 6d are curved. It is also conceivable if one or more resonator mirrors is / are a chirped mirror. Thus, an improved dispersion compensation in the resonator 4 be achieved. It may be advantageous to have a modelocking element 7 in the resonator 4 provide, for example, a Kerr lens or a saturable absorber.

It is also conceivable that in the resonator 4 no laser active medium 2 is provided, but the laser radiation via a coupling mirror directly into the resonator 4 is coupled (similar to the pump laser radiation P). When in the resonator 4 Circulating radiation may in particular be pulsed laser radiation, in particular short or ultrashort laser pulses.

2 shows in frequency space representation the position of the modes one to that in the resonator 4 circulating optical radiation associated optical frequency comb. The optical frequency comb has a plurality of modes with intensity I (f) whose frequencies in the frequency space f can be described by the formula f _m = m × f _rep + f ₀ , where m is a natural number, f _{rep is} a mode spacing and f ₀ are an offset frequency. The position of the modes of the frequency comb is therefore determined by the two parameters offset frequency f ₀ and mode spacing f _rep .

In or on the resonator 4 can be one or more actuators 8a . 8b . 8c . 8d . 8e be provided, by means of which the position of the modes of the frequency comb are adjustable. An actuator may be, for example, a device for adjusting the resonator length, in particular for displacing a resonator end mirror 6a along the optical axis 3 of the resonator 4 , act, in particular a mechanical stepper motor 8a , a piezoelectric motor 8b and / or an electro-optic modulator (EOM). For the present invention, it has proved to be particularly advantageous for setting the resonator length a plurality of actuators with different To provide adjustment accuracy and a different adjustment range. This allows, depending on the situation, either over a large area or with increased accuracy set. This is to move the Resonatorendspiegels 6a a mechanical stepper motor 8a and a piezoelectric motor 8b provided, wherein the adjustment pitch of the piezoelectric motor 8b is finer than that of the mechanical stepper motor 8a , It is also conceivable a piezoelectric motor 8b and to provide an electro-optic modulator, wherein the adjusting step size of the electro-optic modulator is finer than that of the piezoelectric motor 8b ,

Further conceivable actuators, which may alternatively or additionally be provided, are a device for introducing an optical prism 9 in the beam path of the resonator 4 , This in turn can be achieved by a mechanical stepper motor 8a and / or a piezoelectric motor 8b be accomplished. By introducing the optical prism 9 in the beam path of the resonator 4 , or by shifting the position of the prism 9 in a direction perpendicular to the optical axis 3 of the resonator 4 , both the mode spacing f _{m of} the frequency comb (via effects of the prism 9 on the repetition rate) as well as the offset frequency f ₀ (via dispersive effects of the prism 9 ) change.

To change the offset frequency f ₀ , it is also conceivable as a tilting device actuator 8e for an end-mirror 6b of the resonator 4 provided. This may be the end mirror 6b for example, one to the optical axis 3 of the resonator 4 be tilted perpendicular axis.

An example of one not directly on or in the resonator 4 provided actuator is a power controller 8d for the performance of the pump laser 5 , By changing the pump power can be about particular nonlinear intensity dependencies of dispersive properties in the resonator 4 , especially in the laser-active medium 2 to adjust the position of the frequencies of the frequency comb.

In 1 are illustrative actuators for multiple actuators 8a . 8b . 8c . 8d . 8e shown. But it can also be provided a smaller number of actuators, for example, one, two or three. The provision of other actuators is conceivable.

According to the invention, at least one signal S1, S2, S3, S4 is determined, which correlates with an actual value of a degree of freedom F of the frequency comb. In the context of the present invention, the degree of freedom may be any linear combination of the offset frequency f ₀ and the mode spacing f _{rep of} the optical frequency comb. In particular, it may be advantageous if the degree of freedom corresponds to the mode spacing f _rep or the offset frequency f ₀ and thus the signal correlates with an actual value of the mode spacing f _rep or the offset frequency f ₀ .

In the case that the degree of freedom F corresponds to the mode spacing f _{rep of} the frequency comb, the at least one signal S1, S2, S3, S4 can be determined by evaluating a beat of adjacent modes of the frequency comb. For this purpose, for example, the number of oscillations of the beat signal can be determined in a specific time interval by means of a photodetector M1, M2.

In the case that the offset frequency f _{0 is} used as the degree of freedom F, the determination of the at least one signal can be carried out by means of an f: 2f interferometer. In this case, a component of the frequency comb belonging to the optical radiation is frequency doubled and superimposed with a non-frequency doubled component of the optical radiation. The resulting beat has a frequency which corresponds to the offset frequency f _{0 of} the frequency comb, and can be measured by known means.

In 1 two measuring devices M1, M2 are shown, which perform one or more measurements for determining the at least one signal S1, S2, S3, S4. In this case, a corresponding to the frequency comb laser pulse can be measured. It is also conceivable that only one or more than two measuring devices M1, M2 are provided. A the resonator 4 at the Auskoppelspiegel 6b leaving laser pulse is in the in 1 embodiment shown at a first beam splitter D1 divided into two subpulses. One of the subpulses is continued for use as intended. The other subpulse is further split at a second beam splitter D2. The resulting two subpulses of the second generation are each supplied to one of the measuring devices M1, M2. Another positioning of the measuring devices M1, M2 is conceivable, for example in the resonator 4 , The measuring devices M1, M2 may be interferometers or photodiodes or other measuring devices.

As will be explained below, the actual value of the degree of freedom F is regulated to a desired value F _desired . Such a rule is satisfied if the actual value of the degree of freedom F lies within a stabilization range ΔF _{stabilization, which} is described in greater detail below, the setpoint value F _Soll . 3 shows the mutual relationship between the desired value F _Soll and the stabilization range .DELTA.F _{stabilization} and their relation to further value ranges described below the actual value of the degree of freedom F, namely the capture range ΔF _{capture of} the second control unit described below 40 and the validity range .DELTA.F _{validity of} the at least one signal S1, S2, S3, S4.

4 shows a signal processing scheme with which the inventive method can be realized according to an embodiment.

Measured values W1, W2 obtained by the measuring devices M1, M2 can directly represent a signal which correlates with the actual value of the degree of freedom F of the frequency comb. It is also conceivable that the at least one signal S1, S2, S3, S4 results from a measured value W1, W2 by further processing. For example, a measured value W2 may be provided by a further processing unit 20 . 22 . 24 be converted into the signal, in particular by amplifying, normalization, frequency mixing and / or level measurement. As in 4 can be shown from a measured value W2 of the measuring device M2 by respectively different further processing by the further processing units 20 . 22 . 24 several signals S2, S3, S4, which correlate to the actual value of the degree of freedom F. By contrast, the measured value W1 of the measuring device M1 is used directly as signal S1.

According to the invention, it is conceivable that only a single signal S1, S2, S3, S4 correlating with an actual value of the degree of freedom F is determined. Advantageously, however, two, three, four, five or more signals S1, S2, S3, S4 are determined, which correlate with the actual value of the degree of freedom F.

A signal S1 determined by a first measuring device M1 and correlated with the actual value of the degree of freedom F is used as the first input signal to a first control unit 10 transmitted. A first control loop 12 the first control unit 10 receives this first input signal. It is conceivable that further first control circuits 12 are provided, which can also receive this first input signal or alternatively another, correlated with the actual value of the degree of freedom F signal. In the in 4 the embodiment shown is one of a measured value W2 of the second measuring device M2 by the further processing unit 20 derived, further signal S2 as a further first input signal to a further first control loop 12 the first control unit 10 transmitted.

Preferably, the first control loops 12 used sequentially one after the other. The sequential use of the first control loops 12 is preferably by a state machine 30 the first control unit 10 regulated. The state machine 30 can also receive the at least one first input signal and decide depending on at least one first input signal in the sequence of the first control circuits 12 continue to continue. This can be done by the state machine 30 the first control circuits 12 enable or disable, preferably individually. It is advantageous if the state machine 30 all signals from the set of at least one specific signal S1, S2, S3, S4 are supplied.

With the first control unit 10 According to the invention, based on the at least one first input signal, the actual value of the degree of freedom F is in a predetermined _capture range .DELTA.F _{capture of} a second control unit 40 set. The first control circuits have to do this 12 Access to one or more actuators 8a . 8b . 8d , It is conceivable that the first control circuits 12 all the same actuator 8a . 8b . 8d drive. Alternatively, the first control loops 12 different actuators 8a . 8b . 8d to access different setting ranges. In the in 4 shown embodiment, the two first control loops 12 on the stepper motor 8a for shifting the Resonatorendspiegels 6a to.

Especially immediately after switching on the frequency comb, it may happen that the actual value of the degree of freedom F outside of a specified validity range .DELTA.F _{validity of} a signal from the amount of at least one specific signal S1, S2, S3, S4. The validity range ΔF _validity can be different for each signal S1, S2, S3, S4 and the value range of the actual value, in which the signal S1, S2, S3, S4 correlates with the actual value of the degree of freedom F, that there is a one-to-one relationship between the signal S1, S2, S3, S4 and the actual value. In other words, the validity range ΔF _{validity is} the range in which the actual value is correctly determined by determining the signal S1, S2, S3, S4.

It can be checked whether the actual value is within the validity range ΔF _validity . This can be done, for example, by evaluating one of the signals S1, S2, S3, S4 which correlate to the actual value of the degree of freedom F by the state machine 30 respectively. If, for example, a beat of two successive frequency comb modes is provided by the second measuring device M2 and converted into an electrical signal, the further processing unit can 22 be designed as a beat analyzer. By this, the electric beat signal is first passed through a frequency filter which is permeable only to frequencies in a certain range, and then the power level behind the frequency filter as the signal S3 to the state machine 30 passed. This can then determine on the basis of the power level whether the actual value of the degree of freedom F, for example of the mode spacing f _rep , is within the validity range ΔF _{validity of} a signal S1, S2, S3, S4.

If the actual value outside the specified validity range .DELTA.F _validity of a first input signal for driving an actuator 8a . 8b . 8c . 8d . 8e If the signal S1, S2 is present, the adjustment operation is effected by the first control circuits using the relevant signal S1, S2 12 temporarily suspended.

If the actual value for all signals S1, S2 used as the first input signals is outside the valid range .DELTA.F in _question , all first control loops become 12 deactivated and the setting value of at least one actuator (eg the stepping motor 8a in 4 ) stochastically or evenly changed until the validity range ΔF _validity is reached again to return to the range of reliable measurements and thus a meaningful setting by the first control loops 12 get. The state machine communicates for this purpose 30 with a function generator G, which depends on an input of the state machine 30 controls the stochastic or even change in the actuator control value. After the state machine 30 Information about previous signals S1, S2, S3, S4 and actuator settings, the state machine 30 activate the function generator G as a function of previous states in order to get as quickly as possible back into the validity range ΔF _{validity of} at least one signal S1, S2 used as the first input signal.

Based on at least one of the state machine 30 provided, with an actual value of the degree of freedom F correlating signal S1, S2, S3, S4 determines the state machine 30 whether the actual value in the _capture range .DELTA.F _{capture of} the second control unit 40 lies. As in 3 shown, the _capture range .DELTA.F _{capture of} the second control unit 40 are within the scope .DELTA.F _validity, the set value F _set within the scope .DELTA.F _validity and the capture .DELTA.F is _capture, preferably substantially centered. The _capture range ΔF _capture can be usefully determined as the range in which the second control unit 40 the actual value to the setpoint F _setpoint can regulate. For this purpose, the desired value F _{Soll is} within the _capture range .DELTA.F _capture , preferably centered. A regulation of the actual value to the setpoint value F _setpoint is fulfilled according to the invention when the actual value is in a stabilization range ΔF _{stabilization} containing the setpoint value, in particular centered around the desired value F _setpoint , a subrange of the _capture range ΔF _capture . The stabilization range ΔF _{stabilization} may allow deviations of the actual value from the desired value F _setpoint by less than 1%, less than 3%, less than 5% or less than 10% of the setpoint value F _setpoint . Alternatively, the stabilization range ΔF _{stabilization may} allow a deviation of the actual value from the desired value F _setpoint by only a few Hz, for example by at most 2 Hz, 5 Hz, 10 Hz, 20 Hz or 50 Hz.

If through the state machine 30 it is determined that the actual value within the _capture range .DELTA.F _{capture of} the second control unit 40 is the second control unit 40 activated, preferably by the state machine 30 , Then the actual value is calculated using the second control unit 40 to the setpoint F _setpoint . For this purpose, the second control unit 40 one or more second control circuits 42 on. These control one or more actuators 8a . 8b . 8c . 8d . 8e for influencing the actual value. The second control unit 40 can act on actuators 8a . 8b . 8c . 8d . 8e access, also from the first control unit 10 be used. This is in the in 4 shown embodiment of the stepper motor 8a the case. Additionally or alternatively, the second control loops 42 one or more actuators 8a . 8b . 8c . 8d . 8e control that from the first control unit 10 Not used. This is in 4 the case for the piezoelectric motor 8b and the power regulator 8d for the pump laser radiation P. Of course, other or other actuators can be controlled, such. B. an electro-optical modulator 8c or a tilting device 8e for a resonator mirror 6a - 6d ,

The second control circuits 42 the second control unit 40 control the corresponding actuators 8a . 8b . 8d based on a signal S4 transmitted as a second input signal from the set of the at least one signal S1, S2, S3, S4, which correlates with an actual value of the degree of freedom F. It is also conceivable that the second control unit 40 a plurality of second input signals are supplied, based on which the second control circuits 42 the actuators 8a . 8b . 8c . 8d . 8e can control. In the embodiment of 4 is the single second input signal from the measured value W2 of the second measuring device M2 via the further processing unit 24 derived.

The second control circuits 42 are, as in 4 shown, preferably provided cascaded. This is particularly advantageous when the second control circuits 42 each on different actuators 8a . 8b . 8c . 8d . 8e access. In order to optimize the signal usage as possible, it may be advantageous if the first control unit 10 receives at least a second input signal. This can be to one or more first control loops 12 be issued. In particular, the state machine should 30 receive the second input signals. This is particularly advantageous when the state machine 30 the cascading of the second control circuits 42 controlled. This may be the first control unit 10 , in particular the state machine 30 , one or more second control circuits 42 enable or disable, preferably separately.

On the basis of the available data, in particular the first or second input signals, the state machine 30 judge whether a state of successful stabilization of the frequency comb has been achieved. This is the case when the actual value of the degree of freedom F coincides with the desired value F _Soll or lies in the stabilization range ΔF _{stabilization} . Then the state machine can 30 Display or pass on a message indicating the state of successful stabilization. For example, such a message may appear on a screen 80 be issued or read into a computer.

The state machine 30 and none, one or more of the first control loops 12 and / or the function generator G may be implemented as a computer executable program, preferably stored on a computer readable medium.

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• DE 19911103 A1 [0008, 0008]
• DE 10044404 A1 [0008]

Claims (15)

Method for operating a laser device ( 1 ) comprising the steps of: a) providing an optical frequency comb having a plurality of modes whose frequencies are writable by the formula f _m = m × f _rep + f ₀ , where m is a natural number, f _{rep is} a mode spacing, and f ₀ an offset frequency is, b) determining at least one signal (S1, S2, S3, S4) which correlates with an actual value of a degree of freedom (F), wherein the degree of freedom (F) is a linear combination of the offset frequency f ₀ and the mode spacing f _rep is, c) to a first control unit ( 10 ) Transmitting at least one first input signal, the at least one first input signal being contained in the set of the at least one signal (S1, S2, S3, S4), d) with the first control unit ( 10 ) based on the at least one first input signal setting the actual value of the degree of freedom (F) in a predetermined _{capture range} (.DELTA.F _capture ) of a second control unit ( 40 ), e) activating the second control unit ( 40 ) as soon as the _{capture region} (ΔF _capture ) of the second control unit ( 40 ), and f) regulating the actual value with the aid of the second control unit ( 40 ) to a setpoint F _Soll . Method according to Claim 1, characterized in that the first control unit ( 10 ) several first control circuits ( 12 ), Which are sequentially used one by one to (the actual value of the degree of freedom (F) in the capture .DELTA.F _capture) of the second control unit ( 40 ) bring to. Method according to Claim 2, characterized in that the sequential use of the first control circuits ( 12 ) by a state machine ( 30 ) is regulated. Method according to one of the preceding claims, characterized in that the second control unit ( 40 ) a single second control loop ( 42 ) or a group of cascaded second control loops ( 42 ) based on a single second input signal contained in the set of at least one signal ( 53 ) one or more actuators ( 8a . 8b . 8c . 8d . 8e ), whereby the second control circuits connected downstream in the cascading ( 42 ) are adapted to the respective upstream second control circuits ( 42 ) in their respective control areas. A method according to claim 4, characterized in that the second input signal is derived from a beat signal, and that the sign of the beat frequency before or during the activation of the second control unit ( 40 ) is determined. A method according to claim 4 or 5, characterized in that the at least one first input signal comprises the second input signal. Method according to claim 4, 5 or 6, characterized in that the first control unit ( 10 ) one or more second control circuits ( 42 ) separately optionally activated. Method according to one of claims 4 to 7, characterized in that the first control unit ( 10 ) one or more of the second control circuits ( 42 ) controlled actuators ( 8th ) in parallel or alternatively to the second control unit ( 42 ). Method according to one of the preceding claims, characterized by driving one or more actuators ( 8th ) independent of the second control unit ( 42 ) by the first control unit ( 10 ). Method according to one of the preceding claims, characterized in that the first control unit ( 10 ) processes at least one actuator signal which determines the manipulated variable of an actuator ( 8th ). Method according to one of the preceding claims, characterized in that the first control unit ( 10 ), if the actual value lies outside of a defined validity range (ΔF _validity ), a manipulated value of at least one actuator ( 8th ) is changed stochastically or evenly until the validity range (ΔF _validity ) is reached again. A method according to claim 11, characterized in that a check whether the actual value is within the validity range (.DELTA.F _validity ) is made by determining a power level of a beat signal. Method according to claim 11 or 12, characterized in that the at least one actuator ( 8th ) depending on one or more preceding states of the first control unit ( 10 ) is driven. Method according to one of the preceding claims, characterized in that the offset frequency (f ₀ ) and the mode spacing (f _rep ) of the frequency comb are stabilized. A method according to claim 14, characterized by detecting a condition of successful stabilization and then displaying or relaying a message indicating the state of successful stabilization.

Images